Replace the four capture modes with item/track/master scope actions. Each infers
its range (razor-else-time) and enforces FX scope via non-destructive
FX-bypass-around-render (RAII I_FXEN snapshot/restore over ancestors + master).
Corrects the defect of items captured through parent FX.
New pure mode_switch module (header-rect -> N equal segment rects + point hit-test, unit-tested) plus a segmented switch in the bank_panel header: one lit segment per registered mode, click activates via view::applyMode. Grid offset below the header.
Four wet capture actions route to OfflineRenderBackend (snapshot/restore
RENDER_*, 32-bit float), produce a Sample, add to the bank, persist. Pure
render_settings maps source mode to RENDER_SETTINGS and parses P_RAZOREDITS
(tested). No arrange insertion. True dry deferred to M10.
InsertMedia-driven placement reading the bank panel selection, undo-wrapped.
Native length by default; conform-to-tempo is an explicit second action, never
a silent stretch. Pure mode-bit logic (insert_plan) unit-tested.
Register toggle/activate/tag/untag/show-both actions driving the D2
shell and D3 model; reapply saved active mode on project load via a
persist load-signal seam. Shared guidString helper; pure nextModeId
unit-tested.
Docked SWELL window toggled by a new action, drawing the current bank as
per-sample min/max thumbnails (PCM_source + peaks) with an in-memory cache.
Pure grid-layout/cache-key math in bank_grid (tested). Renames peaks::Sample
-> AudioSample to avoid colliding with bank_model::Sample.
Reads I_FOLDERDEPTH into a FolderTree (pure view_tree helper, unit-tested),
snapshots to-be-parked tracks before parking, runs the D1 planner, and applies
park/restore flag + per-FX-offline writes under an undo block. Master and
mute/solo untouched; acts only on tagged tracks.
Serialize BankIndex to project ext state ('reasampler'), reload on project
load, resolve paths project-relative. Save-As copies the bank to the new .rpp;
identity keyed off a minted GUID (not the recycled ReaProject*) so project
switches don't clobber banks. Pure classifyProjectTransition tested.
ICaptureBackend/CaptureRequest seam + OfflineRenderBackend driving snapshotted
RENDER_* with dither/normalize forced off for bit-identical output, RenderFailed
via filesystem check, and a pure capture_paths lib with tests. Spike action registered.
New STATIC lib + peaks_tests (CTest), mirroring bank_model. Float samples,
per-channel (no fold); exact integer bin spans handle remainder, short
buffers, and degenerate inputs with no OOB. Bin-span math guarded against
size_t overflow for pathological binCount.
daniel